Projection of cancer risks from the Japanese atomic bomb survivors to the England and Wales population taking into account uncertainty in risk parameters

Abstract 

Generalized relative risk models, with adjustments to the relative risk for time after exposure and age at exposure and incorporating a linear-quadratic dose response, were fitted to the latest (Life Span Study Report 12) Japanese atomic bomb survivor cancer mortality data using Bayesian Markov Chain Monte Carlo methods, taking account of random errors in the DS86 dose estimates. The resulting uncertainty distributions in the relative risk model parameters were used to derive uncertainties in population cancer risks for a current UK population. Following an assumed administered dose of 1 Sv, leukaemia mortality risks were estimated to be 1.93×10^–2 Sv^–1 (95% CI 1.14, 3.38), or 0.44 years of life lost Sv^–1 (95% CI 0.22, 0.94). Following an assumed administered dose of 1 Sv, solid cancer mortality risks were calculated to be 10.36×10^–2 Sv^–1 (95% CI 8.41, 12.42), or 1.38 years of life lost Sv^–1 (95% CI 1.11, 1.68). In general, solid cancer risks were very similar to those predicted by classical likelihood-based methods; however, leukaemia risks were somewhat higher, by 10–35%, than those predicted by classical likelihood-based methods. This is so in both cases, irrespective of whether or not adjustments are made in these likelihood-based fits for the effects of measurement errors, and the discrepancy for leukaemia tends to be greater at higher doses. Overall, cancer risks predicted by Bayesian Markov Chain Monte Carlo methods are similar to those derived by classical likelihood-based methods and which form the basis of established estimates of radiation-induced cancer risk.